Secure Numerical Computations Using Fully Homomorphic Encryption

Explore secure numerical computations using fully homomorphic encryption in this 33-minute conference talk by Michael Schlottke-Lakemper and Arseniy Kholod. Gain insights into mitigating data exposure risks when processing sensitive information on cloud systems. Learn about two Julia packages, OpenFHE.jl and SecureArithmetic.jl, designed to lower the threshold for using modern, high-performance homomorphic encryption implementations. Discover fundamental concepts of homomorphic encryption, focusing on the CKKS scheme for approximate computations with real numbers. Follow demonstrations on implementing homomorphic encryption in Julia, and explore examples of secure algorithms, including numerical simulations of partial differential equations. Discuss challenges and practical limitations of working with homomorphic encryption, such as performance and algorithmic complexity. Suitable for computational scientists and research software engineers interested in secure numerical computations with Julia, no prior knowledge of homomorphic encryption required.